More than 100,000 babies are born very preterm (?32 weeks gestational age) every year in the United States. Up to 35% develop cognitive deficits and behavioral/psychiatric abnormalities. Unfortunately, these deficits and abnormalities cannot be reliably diagnosed until 3 to 5 years of age. These first few years are now recognized as critical for neuroplasticity, when proven infant stimulation programs could help to restore cognitive function. Our group has identified, in advanced MRI scans obtained at term corrected age (CA), multiple novel biomarkers that appear promising as early predictors of long-term cognitive deficits. Among these, the microstructural volume of diffuse white matter abnormalities (DWMA), as quantified using our novel automated software, is the most promising. We were the first to recognize an association between DWMA and neonatal inflammation-initiating illnesses. Yet, there remains continued debate about the pathological significance of DWMA. Our overall objective is to accurately quantify DWMA in order to elucidate its mechanism and effects on brain connectivity, and then to develop a robust prognostic model at term CA for early childhood cognitive deficits. We propose a large prospective cohort study in very preterm infants with MRI (structural, diffusion, and resting-state functional MRI) at term CA and serial cognitive and behavioral assessments up to 3 years CA. Our central hypothesis is that DWMA is pathological and, when combined with early advanced MRI biomarkers, will accurately predict early childhood cognitive deficits. If confirmed, this would reduce the time to classify individual infans as high- or low-risk by up to 3 years, thus increasing the time available for targeted early intervention and novel neuroprotective therapeutic trials. The three specific aims to test the central hypothesis are: (1) Identify antecedent clinical risk factors for development of DWMA; (2) Determine the association between DWMA and structural and functional brain connectivity; and (3) Predict cognitive deficits using DWMA and advanced MRI biomarkers at term CA. Under the first aim, inflammation-initiating illnesses and other clinical risk factors will be associated wit DWMA. For the second aim, we will examine novel measures of structural and functional connectivity from diffusion MRI and resting state functional MRI and correlate them with DWMA microstructural volume at term CA. Under the third aim, we will combine DWMA with other promising biomarkers using multivariable regression models to robustly predict cognitive and behavioral development. This approach is innovative because it will integrate established epidemiologic principles with advances in neuroimaging, biomedical engineering, and neuropsychology to elucidate DWMA pathology and to develop early prognostic models of cognitive deficits in a large geographically defined population of very preterm infants. The proposed research is significant because it will reduce the time to diagnosis of cognitive and behavioral abnormalities by up to 3 years so that proven early intervention and bio- logically based therapies can be targeted to the highest risk infants, during a period of optimal neuroplasticity.